Unlock instant, AI-driven research and patent intelligence for your innovation.

Image processing apparatus and image processing method

a technology of image processing and image processing apparatus, which is applied in the field of image processing apparatus and image processing method, can solve the problems of difficult to prepare the photographed image of an object, requires a very cumbersome adjustment by an operator, and is difficult to prepare a sphere having the same reflection characteristic as that of the object, so as to achieve the effect of easy generation

Active Publication Date: 2016-11-01
CANON KK
View PDF9 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Benefits of technology

The patent text describes how it's possible to generate a training image for an object that mimics its surface brightness based on information captured from the environment. This can be useful in identifying objects in a variety of situations.

Problems solved by technology

It is, however, difficult to prepare the photographed image of an object in such an orientation as a training image.
Generally, edge extraction processing greatly varies depending on the material of an object, the influence of ambient light, and the like, and requires very cumbersome adjustment by an operator.
It is therefore difficult to prepare a sphere having the same reflection characteristic as that of the object.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Image processing apparatus and image processing method
  • Image processing apparatus and image processing method
  • Image processing apparatus and image processing method

Examples

Experimental program
Comparison scheme
Effect test

first embodiment

[0031]Image processing of generating a training image to be used to generate a dictionary which is referred to in image recognition processing of detecting an object from an input image will be explained below. The training image is generated to approximate the surface luminance of an object to be detected (to be referred to as a “target object” hereinafter) by reflecting environmental conditions based on information obtained by capturing the target object in an actual environment.

[0032][Generation of Training Image]

[0033]FIG. 1A shows the arrangement of an image processing apparatus which performs image recognition processing in the first embodiment.

[0034]A model setting unit 1010 sets the model of a target object and stores it in a model storage unit 1020. An image acquisition unit 1110 acquires a pre-acquired image by capturing the target object, and stores the pre-acquired image in an image storage unit 1120. An observation data distribution obtaining unit 1130 obtains the obser...

embodiment

Modification of Embodiment

[0090]The first embodiment has described an example in which a luminance image and range image including the target object 400 are acquired from capturing by the image capturing apparatus 300. However, the present invention is applicable to even a case in which the image capturing apparatus 300 does not have the distance measurement function. A modification when the image capturing apparatus 300 cannot acquire a range image will be explained.

[0091]FIG. 1B shows the arrangement of an image processing apparatus when a range image is generated by estimating the position and orientation of the target object 400 by using a tracking technique. Note that the same reference numerals as those in FIG. 1A denote the same parts, and a description thereof will not be repeated.

[0092]The arrangement shown in FIG. 1B includes a range image generation unit 1140 which generates a range image. The user superposes the model of the target object 400 on a luminance image capture...

second embodiment

[0093]The second embodiment according to the present invention will be described. The first embodiment has been described on the premise that the target object is an object of a single color. However, the target object sometimes has a plurality of colors. For example, part of a target object is made of black plastic and another part is made of white plastic. In this case, the luminance characteristic changes depending on the portion of the target object. The second embodiment will explain an example in which a training image is generated for a target object having a plurality of luminance characteristics.

[0094]The basic arrangement for performing image recognition processing in the second embodiment is the same as that of FIG. 1A in the first embodiment, except for details of processing by a luminance estimation unit 1210 which estimates the surface luminance of a target object from an observation data distribution.

[0095]FIG. 8 shows the arrangement of the luminance estimation unit ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

In a case where generating a training image of an object to be used to generate a dictionary to be referred to in image recognition processing of detecting the object from an input image, model information of an object to be detected is set, and a luminance image of the object and a range image are input. The luminance distribution of the surface of the object is estimated based on the luminance image and the range image, and the training image of the object is generated based on the model information and the luminance distribution.

Description

BACKGROUND OF THE INVENTION[0001]1. Field of the Invention[0002]The present invention relates to image processing of generating a training image for generating a dictionary to be used in image recognition processing of detecting an object from an input image.[0003]2. Description of the Related Art[0004]Various kinds of research and development have been carried out for image recognition of detecting the image of an object to be detected from an image obtained by capturing objects. The image recognition technique is applied to various fields and used for many actual problems of, for example, face recognition and part recognition in a factory.[0005]This image recognition can be considered from the viewpoint of pattern recognition. In the pattern recognition as well, research has been conducted on classifiers, that is, how to perform classification of input information. There have been proposed various methods such as a neural network, support vector machine (SVM), and randomized trees...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(United States)
IPC IPC(8): G06K9/62G06K9/00G06K9/46G06V10/60G06V10/772
CPCG06K9/6255G06K9/00208G06K9/4661G06K2209/19G06T7/74G06T7/75G06V20/647G06V10/60G06V2201/06G06V10/772G06F18/28B25J9/1697G06T7/0004G06T2207/20081
Inventor AOBA, MASATO
Owner CANON KK